0

Vf Thick Fog

Single-turn 2D grid search environment with noisy L1-distance probes under thick fog, designed for reasoning and efficient tool-use evaluation.

Domain
rl-env
License
mit
Published
Oct 2025

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
prime-hub
Attribution policy →

Top score 272.11 by GPT-5 Mini - 2 models reporting (2 frontier)

Score history

2
075150225300Aug 25GPT-5 Mini

Top models

2
Vf Thick FogBar chart with 2 bars. Highest value: GPT-5 Mini at 272.1.
2 models

Related tools

1
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

FAQ

What is Vf Thick Fog?
Single-turn 2D grid search environment with noisy L1-distance probes under thick fog, designed for reasoning and efficient tool-use evaluation.
What is the current top score on Vf Thick Fog?
The top reported score is 272.11 by GPT-5 Mini, across 2 models reporting (2 from frontier labs).
How can a model improve its Vf Thick Fog score?
Tools linked to Vf Thick Fog on Sophon include Thick FOG RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
What license is Vf Thick Fog under?
Vf Thick Fog is available under mit.